Generally vendors do not understand their products.
The most important use for a platform is frequently something the vendor never considered in either their software design, their marketing plan, or their business model. Or they considered it, and wanted to hold it hostage in the name of an ongoing revenue stream. Consider that most consoles are loss-leaders, with the money being made up on the licensing rights to put software on the box, with per unit royalties to ensure an ongoing revenue stream. The restrictions necessary to implement this business model disable a lot of the potential of the platform as a basis for other applications.
Most QOS algorithms are implemented incorrectly, including those included by default on FreeBSD , Linux, and other systems claiming to support QOS guarantees.
The problem is that these routers are at the border of a very high speed network (the local network), a slower network (your broadband connection), and another high speed network (the actual Internet.
As a result, what happens is that the packets end up being buffered at the router on the other side of the border router, or in the computer (if in the home, and your only router is the border router).
This effectively starves the connection of packet buffers on one side or the other, if there is a low priority but high bandwidth demand application sitting on the other end. No matter what you do in terms of rate based algorithms, your border router is not going to be able to do anything about the starvation, which is not happening on a box under its direct control.
The only approach that works very well (and which most traffic shaping software does not implement) is to advertise a smaller TCP window to the other end of the link, so that there are buffers available because the transmitting end believes that the TCP window is full on the receiving end of the TCP connection. RED queueing can also work, but only protects the endpoint equipment, and ignores the problems of the intermediate routers; typically, this makes things much worse for the intermediate hop routers, which have to put up with retransmits of the same data.
I've come to the conclusion that this is, in fact, why the broadband companies, which have fast links to their head-ends, but have relatively slower links to residences, have decided to start a war against peer-to-peer technology, and why some are suggesting higher feeds for streaming video users by placing bandwidth caps and punishing users who exceed them: you effectively starve the other customers behind the same head end as yourself by using all the head end router buffers for traffic that hasn't gotten to you yet, or you flood outbound with retransmits for unack'ed data.
My suggestion would be to buy something that know how to traffic shape with TCP windows (and it's my advice to both the poster and the infrastructure companies, who have apparently failed to grasp the idea that traffic shaping is not a job for amatuers who only prioritize outbound traffic, like AltQ and friends).
Mac OS X is the *first* UNIX(tm) *not* derived from AT&T sources.
I was one of the people who made Mac OS X into UNIX(tm), and we started from not even being able to compile the test suite.
My first one line header file change to xnu to test the water (not defining size_t in ) broke 156 projects, including Open Source that was written by people who assumed promiscuous #include files, in violation of the standard.
A relatively small team of us fixed well over 40,000 total test case failures in a period of about 2.5 years, many of those in command line tools, most of that code being pushed back out to the various Open Source projects. Like, oh, "gcc", "bash", "vim", "tar", "bc", "pax", and hundreds of others, which are now UNIX conformant because of us.
In the middle of things we were working 80 hour weeks, sometimes more.
At the end... *almost no one noticed the changes*, because we worked our *asses* off to make sure there was so close to zero *both binary and source* compatibility issues that it would *not* be noticed. One member of the team put it this way: "It's like raising everyone 12 feet into the air, and replacing the Earth underneath them, then lowering them back down to the ground".
All told, we changed more lines of code in the kernel, libraries, compiler, and UNIX(tm) standardized utilities, than all of the non-conformance related changes in Tiger and Leopard combined. I counted.
And then we published the sources for everything needed to build your own Darwin system that could pass the UNIX(tm) conformance test, including our kernel.
So let me repeat: you, sir, do not know what you are talking about.
(1) Point the first: (I)ISPs who have common carrier status:
(a) provide DSL services over phone lines
(b) provide VOIP and other telephony over Internet services
(II) ISPs have been held to fall under common carrier statutes by court cases:
(a) 5th Circuit Court of Appeals
(b) 9th Circuit Court of Appeals
(III) ISPs *should* have common carrier status
(a) to prevent statutory requirement that are technologically impossible to enforce
(b) to prevent the ISPs from acting as anything other than "dumb packet pipes"
(c) to preclude lobbies from taking over the Internet, they way they have Congress
(2) Point the second: (I) Usenet is a store-and-forward network architecture
(II) Usenet posts can be blocked/cancel-botted in jurisdictions where child pornography is illegal (and should be)
(III) Undesirable web sites can be blocked at the border routers (ask China)
(IV) It is cheaper and more effective to block at the border
(V) The only reason they are going after ISPs is that it is easier to get them to cave to pressure
(VI) An Internet where any church, civic group, political party, cult, or school board could control content by putting financial pressure on an ISP, would suck
Do not get me wrong; I am all for hanging child pornographers.
However, the precedents being set by this are terrible.
One precedent being set here is that ISPs are taking responsiblity for Internet content transiting their backbone. This has both short and long term consequences. The short term consequence is a chilling effect on freedom of speech, due to removal of effective common carrier status from these ISPs. They are now subject to various forms of censorship by government agencies. Longer term, I expect that there will be law suits against ISPs that fail to block "objectionable content", be it child pornography, hate speech, screeds against or for select religions, or, eventually, political speech.
The second precedent is far, far more troubling. The article states:
The agreement is designed to bar access to Web sites that feature child pornography by requiring service providers to check against a registry of explicit sites maintained by the Center for Missing and Exploited Children.
This essentally means that control over Internet content obtainable through these ISPs is being given over to a private watchdog organization, with no public oversight. A Star Chamber. Further, and an even more distressing implication, is that the Center for Missing and Exploited Children knows about these sites, has known about these sites, and is they are not being shut down. I can reach only one of two conclusions from this:
(A) The Center for Missing and Exploited Children is allowing these sites to persist as a political whipping boy, in order to seize editorial control over the content of the Internet, and the children being exploited by these sites are considered "necessary losses" in service of the "greater good"; in other words, the children are being exploited by the Center for Missing and Exploited Children, as well as the pornographers.
(B) The Center for Missing and Exploited Children can not prove legally that these sites are in fact purveying child pornography, which would get them taken down and their operators prosecuted under existing anti child pornography laws, so instead they are engaging in illegal tactics under the RICO and Sherman Antitrust Acts to get them censored.
Either way, I do not like a private police force obtaining this type of power, no matter how noble their intentions; I'm sure that the Pinkerton agents involved in putting down the Homestead Strike felt that their cause was just and noble as well - and look how that turned out.
Your stance on guns is clearly uninformed by living in a country that owes its existence to the successful violent overthrow of an oppressive and non-representative government, by an armed citizenry, using weapons comparable to those available to the government which it was overthrowing.
AKA the United States.
"Those who do not learn from history are doomed to repeat it" - George Santayana
I'm still stuck at the technological hurdle of actually being able to _implement_ such filters in the first place, given that it's an NP-incomplete problem.
It's all well and good to scream "protect the children!" at the top of your lungs, but what technology are you proposing to identify and interdict obscene content?
I am a little more concerned with my car being driven off a cliff by an automated traffic control system because some asshat decides to invade some other asshat, and to hell with the civilians using the system.
The insistence on a NAVWAR backdoor is rather stupid. In the last three wars in which it has been involved, the U.S. has pretty much had its rear kicked by enemies using what amounts to 1940s technology. The danger to US troops is not from WMDs, it's from IEDs made in peoples kitchens using easily obtained ordinance, generally with U.S. serial numbers on it.
If they want to blow me up, they're going to do it by setting up a bomb that reacts to the RFID in my "Real ID" card, U.S. Passport, or the pressure sensors in my tires, all of which are government mandated, and all of which go where I go, and so are really useful for targeting me both generally ("look, and American!") or personally. Or they'll use my IMEI on my cell phone, which on differs in that I'm not required to carry it, but probably will anyway.
If someone can build a missile that can get to me from where they are, then unless I am sitting in a bunker, a few hundred feet for going inertial or using airport beacons instead of GPS isn't going to matter much one way or another.
Diablo Canyo powers 2.2 million people with two reactors, so you are talking 17 more installations of a comparable size to power California.
I'm pro-nuclear, and I can't see that happening in California, even if the price of natural gas goes up at the California/Nevada border again, as it did under Enron. California is all about NIMBY. Now build them in some other state and run wires, and California would likely love the idea.
Which, assuming high winds, will provide about 1/3 the power output of one of the Diablo Canyon reactors. Their own estimates are closer to 1/6 that load on average. That works out to being able to supply power for about 180,000 people (Diablo Canyon's two reactors supply for about 2.2 million homes).
To put this in perspective, all the wind power generating capacity currently deployed in California is about 3/4 of one reactor at Diablo Canyon, and that's assuming the wind is blowing constantly at the average, or about 2.5 times what Cape Wind plans on deploying, if it can get regulatory approval, and prove negligible environment impact from the construction and deployment both.
That isn't a small amount of generating capacity, but the fact that this is going to take building 130 generating stations to achieve, and a huge area (as you pointed out: not chump change, with regard to ocean acreage). It's also going to only end up supplying about 75% of the overall usage of Cape Cod, and the two islands of Martha's Vinyard and Nantucket - not a lot of people.
To put that figure in perspective, that's 4.5 x 5.4 nautical miles square, or about 30 square non-nautical miles, to supply 135,000 people.
"How about salaries? Why do doctors need so much education to diagnose most problems?"
A lot of Education != Higher Salary; this is the big lie. As someone who is familiar with the collections business, the only people harder to collect from than doctors is dentists. Doctors tend to live ostentatiously - at or above their means - because that's expected of them. Your average GP does NOT make that much money.
"Do you really need all that staff just to take BP readings and chart notes?"
How much do you make an hour? Do you do your own laundry? The hour you spend doing laundry is worth however much you make an hour to pay someone else to do it instead. Doctors cost more per hour than support staff, so I'm pretty happy to have the doctor looking at patients rather than taking the BP and weighing people and getting an initial history and description of the complaint. Just like I'm happy to pay the people in the HR department do engineers don't waste valuable company time interviewing the idiots that HR weeds out before they get that far.
"Distribution of costs. My employer pays into Worker's Compensation so if I get hurt at work, my medical bills are covered. But if the state denies my claim, guess who gets billed? My insurance company."
Worker's Comp is a state-run insurance company for specific types of medical care and disability. So now you are complaining, once again, that the problem isn't insurance costs, it's insurance costs.
"Profits. Private companies need to make profits."
No. The IRS and the Department of State for your state only requires that they "be in business with the intent to make a profit". They only have to net $1 in profit to qualify for that. How do you think we regulate public utilities and rights-of-way for railroads, both of which are run by private companies, but using public resources to implement their services?
"Free care. Hospitals have to subsidize coverage for the uninsured."
Emergency rooms have to treat people when they show up IFF they get state or federal funding. If not, then the typical response is "stabilize them, then turf then to County". County hospitals (state run facilities, paid for by taxes) take the brunt of subsidized/state paid urgent and emergency care, but they do so out of your taxes.
"Waste. Does every bloody specialist need to take their own set of X-rays?"
It depends on what they are looking for. Sometimes, yes. Different exposure levels reveal different information. X-Rays are relatively cheap; it's the liability assumed by the X-Ray technician and the radiologist and the hospital that are what cost you.
"Do you really need that MRI or expensive test?"
Depends on your complaint. MRIs, again are relatively cheap to do in theory, but not so in practice. I worked on the console software for a G.E. Medical Systems MRI back in the late 1980's. The development costs were about 3 times higher than for a similar system not intended to qualify as a "life support system", but the biggest billable expense was the annuity to pay the liability insurance, in case we screwed up and someone died because of it. Same with a blood gas monitor I worked on, and the same for a blood oxygen sensor.
"Or that brand name med?"
Now you are complaining about the patent system and the fact that the 60+5 of the cost of any pharmeceutical goes to marketing. This is Slashdot, I think we can all agree both need reform. However, the answer in most cases is "yes", if your doctor thinks you need it, having MedCo or some other managed care company substitute your allergy medication for a generic that contains the very allergen that the medication was being prescribed to deal with in the first place is pretty stupid.
"Costs of insurance is low on the list, at least for most doctors."
The thing you are missing is that you are paying for medical inssurance, which is paying your doctor, who then turns around and buys equipment from a manufacturer that's paying liability
"Instead of arguing on how to split the outrageous costs, we should actually focus on fixing the problems leading to the high prices in the first place."
You mean, like malpractice insurance for doctors and hospitals, and liability insurance for drug companies, hospitals, and device manufacturers?
So the problem isn't actually the cost of insurance, it's the cost of insurance.
ANY age is tragic when YOU are the one doing the dying.
"Your unjustified assumption.. The fatality occurs at such a young age that it would generally be considered tragic"
It's this type of asinine thinking that would lead a government controlled health insurance program from authorizing the treatment of a disease if you were over a certain age.
It was never promised by engineers. I have no idea who you were talking to. Can you point to a URL where there was a written statement that said it was promised?
It stores 32 bit values in the file ID database; how the hell are you supposed to have 64 bit Carbon applications interoperate with 32 bit legacy applications that are passing around 64 bit values and truncating them down to 32? How are you supposed to maintain binary compatibility with 32 bit applications that have on disk and in-resource values that are expected to be laid out a specific way?
This was a technical decision, and it was the only possible one that could have been made; those structures and data contents were never written with 64 bit in mind. Anyone who is complaining about this obviously understands nothing about the Carbon internals.
Are we sure Creative could add features (such as Vista compatibility) to this card they already shipped, without asking for payment, and not been in trouble over SOX? http://en.wikipedia.org/wiki/Sarbanes-Oxley_Act.
Or pocket the money and not lay any new cables anyway, which is the most likely thing to happen.
The problem with these companies is that the infrastructure should be publicly owned. Then if Comcast doesn't want to provide the service you want, you can fire them and hire someone who will.
The problem comes from municipality granted monopolies for payola from the companies who want the monopolies. The municipality gets a chunk of money that it can spend on something other than the service, and the company gets to wield monopolistic power and get away with providing an inferior service with no fear of the consequences of doing so.
Slashdot Mirror
"$20 per pound is WAY too small." ...
"That's more than $40/kg"
Uh, you do realize a kg is 2.2 pounds, right?
Did you work on the Mars Climate Orbiter mission by any chance?
-- Terry
I was being facetious. Clearly it would be a bad idea.
-- Terry
Actually all this would do, if GPS units were required, is let us know where they died, within 1m of resolution.
Maybe if you mandated people install governors, so that they couldn't accelerate over the speed limit to avoid accidents...
-- Terry
Generally vendors do not understand their products.
The most important use for a platform is frequently something the vendor never considered in either their software design, their marketing plan, or their business model. Or they considered it, and wanted to hold it hostage in the name of an ongoing revenue stream. Consider that most consoles are loss-leaders, with the money being made up on the licensing rights to put software on the box, with per unit royalties to ensure an ongoing revenue stream. The restrictions necessary to implement this business model disable a lot of the potential of the platform as a basis for other applications.
-- Terry
I say we put quotas on Congress, first; talk about your "boys clubs"...
Why don't they get their own house in order?
-- Terry
If you also wondered what the hell ICFP was...
"The ACM SIGPLAN International Conference on Functional Programming (ICFP)"
http://www.icfpconference.org/ ...it's their annual programming contest.
-- Terry
QOS guarantees and border routers
Most QOS algorithms are implemented incorrectly, including those included by default on FreeBSD , Linux, and other systems claiming to support QOS guarantees.
The problem is that these routers are at the border of a very high speed network (the local network), a slower network (your broadband connection), and another high speed network (the actual Internet.
As a result, what happens is that the packets end up being buffered at the router on the other side of the border router, or in the computer (if in the home, and your only router is the border router).
This effectively starves the connection of packet buffers on one side or the other, if there is a low priority but high bandwidth demand application sitting on the other end. No matter what you do in terms of rate based algorithms, your border router is not going to be able to do anything about the starvation, which is not happening on a box under its direct control.
The only approach that works very well (and which most traffic shaping software does not implement) is to advertise a smaller TCP window to the other end of the link, so that there are buffers available because the transmitting end believes that the TCP window is full on the receiving end of the TCP connection. RED queueing can also work, but only protects the endpoint equipment, and ignores the problems of the intermediate routers; typically, this makes things much worse for the intermediate hop routers, which have to put up with retransmits of the same data.
I've come to the conclusion that this is, in fact, why the broadband companies, which have fast links to their head-ends, but have relatively slower links to residences, have decided to start a war against peer-to-peer technology, and why some are suggesting higher feeds for streaming video users by placing bandwidth caps and punishing users who exceed them: you effectively starve the other customers behind the same head end as yourself by using all the head end router buffers for traffic that hasn't gotten to you yet, or you flood outbound with retransmits for unack'ed data.
My suggestion would be to buy something that know how to traffic shape with TCP windows (and it's my advice to both the poster and the infrastructure companies, who have apparently failed to grasp the idea that traffic shaping is not a job for amatuers who only prioritize outbound traffic, like AltQ and friends).
-- Terry
You, sir, do not know what you are talking about.
Mac OS X is the *first* UNIX(tm) *not* derived from AT&T sources.
I was one of the people who made Mac OS X into UNIX(tm), and we started from not even being able to compile the test suite.
My first one line header file change to xnu to test the water (not defining size_t in ) broke 156 projects, including Open Source that was written by people who assumed promiscuous #include files, in violation of the standard.
A relatively small team of us fixed well over 40,000 total test case failures in a period of about 2.5 years, many of those in command line tools, most of that code being pushed back out to the various Open Source projects. Like, oh, "gcc", "bash", "vim", "tar", "bc", "pax", and hundreds of others, which are now UNIX conformant because of us.
In the middle of things we were working 80 hour weeks, sometimes more.
At the end... *almost no one noticed the changes*, because we worked our *asses* off to make sure there was so close to zero *both binary and source* compatibility issues that it would *not* be noticed. One member of the team put it this way: "It's like raising everyone 12 feet into the air, and replacing the Earth underneath them, then lowering them back down to the ground".
All told, we changed more lines of code in the kernel, libraries, compiler, and UNIX(tm) standardized utilities, than all of the non-conformance related changes in Tiger and Leopard combined. I counted.
And then we published the sources for everything needed to build your own Darwin system that could pass the UNIX(tm) conformance test, including our kernel.
So let me repeat: you, sir, do not know what you are talking about.
-- Terry
(1) Point the first:
(I)ISPs who have common carrier status:
(a) provide DSL services over phone lines
(b) provide VOIP and other telephony over Internet services
(II) ISPs have been held to fall under common carrier statutes by court cases:
(a) 5th Circuit Court of Appeals
(b) 9th Circuit Court of Appeals
(III) ISPs *should* have common carrier status
(a) to prevent statutory requirement that are technologically impossible to enforce
(b) to prevent the ISPs from acting as anything other than "dumb packet pipes"
(c) to preclude lobbies from taking over the Internet, they way they have Congress
(2) Point the second:
(I) Usenet is a store-and-forward network architecture
(II) Usenet posts can be blocked/cancel-botted in jurisdictions where child pornography is illegal (and should be)
(III) Undesirable web sites can be blocked at the border routers (ask China)
(IV) It is cheaper and more effective to block at the border
(V) The only reason they are going after ISPs is that it is easier to get them to cave to pressure
(VI) An Internet where any church, civic group, political party, cult, or school board could control content by putting financial pressure on an ISP, would suck
-- Terry
Do not get me wrong; I am all for hanging child pornographers.
However, the precedents being set by this are terrible.
One precedent being set here is that ISPs are taking responsiblity for Internet content transiting their backbone. This has both short and long term consequences. The short term consequence is a chilling effect on freedom of speech, due to removal of effective common carrier status from these ISPs. They are now subject to various forms of censorship by government agencies. Longer term, I expect that there will be law suits against ISPs that fail to block "objectionable content", be it child pornography, hate speech, screeds against or for select religions, or, eventually, political speech.
The second precedent is far, far more troubling. The article states:
This essentally means that control over Internet content obtainable through these ISPs is being given over to a private watchdog organization, with no public oversight. A Star Chamber. Further, and an even more distressing implication, is that the Center for Missing and Exploited Children knows about these sites, has known about these sites, and is they are not being shut down. I can reach only one of two conclusions from this:
(A) The Center for Missing and Exploited Children is allowing these sites to persist as a political whipping boy, in order to seize editorial control over the content of the Internet, and the children being exploited by these sites are considered "necessary losses" in service of the "greater good"; in other words, the children are being exploited by the Center for Missing and Exploited Children, as well as the pornographers.
(B) The Center for Missing and Exploited Children can not prove legally that these sites are in fact purveying child pornography, which would get them taken down and their operators prosecuted under existing anti child pornography laws, so instead they are engaging in illegal tactics under the RICO and Sherman Antitrust Acts to get them censored.
Either way, I do not like a private police force obtaining this type of power, no matter how noble their intentions; I'm sure that the Pinkerton agents involved in putting down the Homestead Strike felt that their cause was just and noble as well - and look how that turned out.
-- Terry
Your stance on guns is clearly uninformed by living in a country that owes its existence to the successful violent overthrow of an oppressive and non-representative government, by an armed citizenry, using weapons comparable to those available to the government which it was overthrowing.
AKA the United States.
"Those who do not learn from history are doomed to repeat it" - George Santayana
-- Terry
Everyone is bitching about filtering...
I'm still stuck at the technological hurdle of actually being able to _implement_ such filters in the first place, given that it's an NP-incomplete problem.
It's all well and good to scream "protect the children!" at the top of your lungs, but what technology are you proposing to identify and interdict obscene content?
-- Terry
If you break this up into tuples of 4 and Google it, you get some interesting matches from geomagnetic observatory data.
-- Terry
I am a little more concerned with my car being driven off a cliff by an automated traffic control system because some asshat decides to invade some other asshat, and to hell with the civilians using the system.
The insistence on a NAVWAR backdoor is rather stupid. In the last three wars in which it has been involved, the U.S. has pretty much had its rear kicked by enemies using what amounts to 1940s technology. The danger to US troops is not from WMDs, it's from IEDs made in peoples kitchens using easily obtained ordinance, generally with U.S. serial numbers on it.
If they want to blow me up, they're going to do it by setting up a bomb that reacts to the RFID in my "Real ID" card, U.S. Passport, or the pressure sensors in my tires, all of which are government mandated, and all of which go where I go, and so are really useful for targeting me both generally ("look, and American!") or personally. Or they'll use my IMEI on my cell phone, which on differs in that I'm not required to carry it, but probably will anyway.
If someone can build a missile that can get to me from where they are, then unless I am sitting in a bunker, a few hundred feet for going inertial or using airport beacons instead of GPS isn't going to matter much one way or another.
-- Terry
Diablo Canyo powers 2.2 million people with two reactors, so you are talking 17 more installations of a comparable size to power California.
I'm pro-nuclear, and I can't see that happening in California, even if the price of natural gas goes up at the California/Nevada border again, as it did under Enron. California is all about NIMBY. Now build them in some other state and run wires, and California would likely love the idea.
-- Terry
Which, assuming high winds, will provide about 1/3 the power output of one of the Diablo Canyon reactors. Their own estimates are closer to 1/6 that load on average. That works out to being able to supply power for about 180,000 people (Diablo Canyon's two reactors supply for about 2.2 million homes).
To put this in perspective, all the wind power generating capacity currently deployed in California is about 3/4 of one reactor at Diablo Canyon, and that's assuming the wind is blowing constantly at the average, or about 2.5 times what Cape Wind plans on deploying, if it can get regulatory approval, and prove negligible environment impact from the construction and deployment both.
That isn't a small amount of generating capacity, but the fact that this is going to take building 130 generating stations to achieve, and a huge area (as you pointed out: not chump change, with regard to ocean acreage). It's also going to only end up supplying about 75% of the overall usage of Cape Cod, and the two islands of Martha's Vinyard and Nantucket - not a lot of people.
To put that figure in perspective, that's 4.5 x 5.4 nautical miles square, or about 30 square non-nautical miles, to supply 135,000 people.
-- Terry
Two words: "opportunity cost"
http://en.wikipedia.org/wiki/Opportunity_cost
The time you sped doing laundry can not be spent doing something else.
-- Terry
"How about salaries? Why do doctors need so much education to diagnose most problems?"
A lot of Education != Higher Salary; this is the big lie. As someone who is familiar with the collections business, the only people harder to collect from than doctors is dentists. Doctors tend to live ostentatiously - at or above their means - because that's expected of them. Your average GP does NOT make that much money.
"Do you really need all that staff just to take BP readings and chart notes?"
How much do you make an hour? Do you do your own laundry? The hour you spend doing laundry is worth however much you make an hour to pay someone else to do it instead. Doctors cost more per hour than support staff, so I'm pretty happy to have the doctor looking at patients rather than taking the BP and weighing people and getting an initial history and description of the complaint. Just like I'm happy to pay the people in the HR department do engineers don't waste valuable company time interviewing the idiots that HR weeds out before they get that far.
"Distribution of costs. My employer pays into Worker's Compensation so if I get hurt at work, my medical bills are covered. But if the state denies my claim, guess who gets billed? My insurance company."
Worker's Comp is a state-run insurance company for specific types of medical care and disability. So now you are complaining, once again, that the problem isn't insurance costs, it's insurance costs.
"Profits. Private companies need to make profits."
No. The IRS and the Department of State for your state only requires that they "be in business with the intent to make a profit". They only have to net $1 in profit to qualify for that. How do you think we regulate public utilities and rights-of-way for railroads, both of which are run by private companies, but using public resources to implement their services?
"Free care. Hospitals have to subsidize coverage for the uninsured."
Emergency rooms have to treat people when they show up IFF they get state or federal funding. If not, then the typical response is "stabilize them, then turf then to County". County hospitals (state run facilities, paid for by taxes) take the brunt of subsidized/state paid urgent and emergency care, but they do so out of your taxes.
"Waste. Does every bloody specialist need to take their own set of X-rays?"
It depends on what they are looking for. Sometimes, yes. Different exposure levels reveal different information. X-Rays are relatively cheap; it's the liability assumed by the X-Ray technician and the radiologist and the hospital that are what cost you.
"Do you really need that MRI or expensive test?"
Depends on your complaint. MRIs, again are relatively cheap to do in theory, but not so in practice. I worked on the console software for a G.E. Medical Systems MRI back in the late 1980's. The development costs were about 3 times higher than for a similar system not intended to qualify as a "life support system", but the biggest billable expense was the annuity to pay the liability insurance, in case we screwed up and someone died because of it. Same with a blood gas monitor I worked on, and the same for a blood oxygen sensor.
"Or that brand name med?"
Now you are complaining about the patent system and the fact that the 60+5 of the cost of any pharmeceutical goes to marketing. This is Slashdot, I think we can all agree both need reform. However, the answer in most cases is "yes", if your doctor thinks you need it, having MedCo or some other managed care company substitute your allergy medication for a generic that contains the very allergen that the medication was being prescribed to deal with in the first place is pretty stupid.
"Costs of insurance is low on the list, at least for most doctors."
The thing you are missing is that you are paying for medical inssurance, which is paying your doctor, who then turns around and buys equipment from a manufacturer that's paying liability
Source of costs of medical care is not insurance?
"Instead of arguing on how to split the outrageous costs, we should actually focus on fixing the problems leading to the high prices in the first place."
You mean, like malpractice insurance for doctors and hospitals, and liability insurance for drug companies, hospitals, and device manufacturers?
So the problem isn't actually the cost of insurance, it's the cost of insurance.
Or am I missing something obvious here?
-- Terry
ANY age is tragic when YOU are the one doing the dying.
.. The fatality occurs at such a young age that it would generally be considered tragic"
"Your unjustified assumption
It's this type of asinine thinking that would lead a government controlled health insurance program from authorizing the treatment of a disease if you were over a certain age.
-- Terry
"We'll have our compensation somehow."
Free Brown Zunes for everyone?
-- Terry
It was never promised by engineers. I have no idea who you were talking to. Can you point to a URL where there was a written statement that said it was promised?
-- Terry
CARBON CAN NOT BE PORTED TO 64 BIT
It stores 32 bit values in the file ID database; how the hell are you supposed to have 64 bit Carbon applications interoperate with 32 bit legacy applications that are passing around 64 bit values and truncating them down to 32? How are you supposed to maintain binary compatibility with 32 bit applications that have on disk and in-resource values that are expected to be laid out a specific way?
This was a technical decision, and it was the only possible one that could have been made; those structures and data contents were never written with 64 bit in mind. Anyone who is complaining about this obviously understands nothing about the Carbon internals.
-- Terry
Or more likely Sarbanes-Oxley.
Are we sure Creative could add features (such as Vista compatibility) to this card they already shipped, without asking for payment, and not been in trouble over SOX? http://en.wikipedia.org/wiki/Sarbanes-Oxley_Act.
-- Terry
Or pocket the money and not lay any new cables anyway, which is the most likely thing to happen.
The problem with these companies is that the infrastructure should be publicly owned. Then if Comcast doesn't want to provide the service you want, you can fire them and hire someone who will.
The problem comes from municipality granted monopolies for payola from the companies who want the monopolies. The municipality gets a chunk of money that it can spend on something other than the service, and the company gets to wield monopolistic power and get away with providing an inferior service with no fear of the consequences of doing so.
-- Terry